Our greatest understanding of underlying mechanisms in animal communication has come from studies of highly specialized systems including electric fish, songbirds, bats, and humans. Noticeably absent from this list are the non-human primates. This proposal seeks funding to explore the mechanisms of vocal production and perception in cotton-top tamarin monkeys, focusing explicitly on the role of auditory feedback in the generation of species-typical signals, as well as the necessary and sufficient features for perceptual classification. We test five hypotheses focusing on the "combination long call" (CLC), a highly specialized signal designed to maintain contact and attract mates; we use four different methods to evaluate these hypotheses: vocal modification in response to changes in environmental acoustics, habituation discrimination, antiphonal calling, and phonotaxis. HI: Vocal production in tamarins is immune to interference, such that when they initiate a CLC, it continues to completion in the face of potentially competing acoustic or visual stimuli. If a competing stimulus causes interruption, production will terminate in between syllables as opposed to within a syllable. H2: Tamarins can use auditory feedback to modulate voice amplitude in response to ongoing increases ambient noise levels, but fail to exhibit such plasticity when changes arise during call production; infants show less plasticity than adults, thereby indicating the role of experience in vocal development. H3: The capacity for individual recognition in tamarins relies on only one syllable type within the CLC (i.e., the whistle), and specifically, a minimum number of whistles as opposed to total energy or syllable rate. H4: Based on comparative acoustic analyses of several other tamarin species, we hypothesize that cotton-top tamarins will show selective sensitivity to the order, duration, and fundamental frequency of syllable types within the CLC, preferentially approaching calls that fall within the natural range of variation for these parameters. H5: Studies reveal that recognition of the CLC is based on a suite of relevant acoustic features. If such features are either entirely absent or significantly perturbed, then processing time will increase, resulting in a longer latency to antiphonally call back to the signal. Results from this project will set the stage for future research on the underlying neurobiological circuitry and the extent to which it is homologous with humans.